DESCRIPTION: Tuberculosis is a major infectious disease which is responsible for three million deaths worldwide each year. The structural knowledge of both antibody and T-cell reactive epitopes of the major tuberculosis antigens is immensely important for the rational design of diagnostic reagents and vaccines. These objectives are justified because currently used diagnostic reagents and the BCG vaccine are not very reliable. Also, there is no data base of structures for such mycobacterial antigens and their complexes with host immune-response proteins. The proposed structural studies have three major objectives: (1) to understand at atomic level the structure and function of species-specific protein antigens isolated from Mycobacterium tuberculosis, an etiological agent of tuberculosis; (2) to study atomic structures of complexes of mycobacterial antigens with Fab or Fv fragments of well characterized monoclonal antibodies to delineate antibody reactive epitopes; (3) to probe further structure-function relationships of antigens by site-directed mutagenesis. The short term goal is to determine well refined high resolution x-ray structures of two species-specific protein antigens, the 38 kD and 16 kD proteins. The 38 kD antigen is a phosphate-binding protein and shares 30 percent amino acid similarity with periplasmic phosphate binding protein from E. coli, whereas, the 16 kD antigen is a small heat shock protein with chaperonin activity and shares 30 percent amino acid identity with the alpha-crystalline family of proteins. Large quantities of recombinant proteins of both types are available for crystalline purposes. The 38 kD antigen has been crystallized in two crystal forms, whereas the 16 kD crystals are not yet large enough to be characterized. The 2 A resolution data from native and low resolution data from heavy atom derivatized crystal of the 38 kD antigen are available for structure determination. The long-term goal is to prepare anti-38 kD and anti-16 kD monoclonal antibody fragments (Fab or Fv) using enzymatic and recombinant methods in order to attempt the crystalliza-tion of antigen-antibody complexes. The knowledge of tertiary structures will be used for structural mapping as well as for determining the conformation of both antibody and T-cell reactive epitopes. The primary structures of peptides constituting such epitopes have been inferred from immunochemical studies.